I think the answer is yes, it is possible

THank you Deb, I will try tomorow.

in fact, I have made a scan with a few problems (session putty closed due to timeout during the scan).

AFter a few tries, I was thinking that everything was ok Because all savesets from the two tapes seems to be there.

But when launching  arecover session, some files and directory where missing.

Now I want to restart from scratch and delete everything from these 2 tapes in Networker databases.

Can you give me some advise to achieve this?

The easiest way is to delete the volumes from the media index using "nsrmm -d ".

Do not forget that you can also pipe the output to a file for later investigations.

It is also important that the save set have not passed their browse date - otherwise NW will delete the file index information right away. To prevent that, correct browse/retention date for the save sets as follows before you start the scan process:

   nsrmm -w -e -S ssid

Because I think my index is corrupted, I want also restart from scratch for this particular client.

To resume:

1. Delete the media index

# nsrmm -d

2. Delete the index file for this client

# ??? need your advises

3. Scan the tape to recover media and index files.

# scanner -i /dev/rmtx.1

Note:

I think the index file is corrupted because after several scan (some aborted), when trying a recover I can see that some files or directory are missing within the list.

Thats why I want to restart from scratch.

The deletion of the media index is not necessary - and it would be difficult for a particular client.

If you really want to do that, you must delete all his save sets (nsrmm -d -S ).

To delete the file index, simply remove/delete the clients index directory, usually stored in "/nsr/index/ ".

To ensure that it has been deleted, run "nsrls

Wort case update the information with "nsrck -X ".

Rerun scanner. Again I recommend you pipe the output to a file for further investigation.

You can also scan a specific save set using "scanner -S ", However, in this case you must know the ssid, of course.

Also make sure that you have the latest NW version installed. I remember an issue with scanner (about 4 years old with NW 7.5?) where the command ran successfully but in the end did not update the index information. This was especially for a NW/Windows server ... but you never know. Again, I recommend creating a logfile for an easy verification.

Thanks,

I have now restarted from scratch and have the following issue reported in a new post:

scanner command ask the same tape twice